Abstract | ||
---|---|---|
Many computer vision algorithms include a robust estimation step where model parameters are computed from a data set containing a significant pro- portion of outliers. The RANSAC algorithm is possibly the most widely used robust estimator in the field of computer vision. In the paper we show that un- der a broad range of conditions, RANSAC efficiency is significantly improved if its hypothesis evaluation step is randomized. A new randomized (hypothesis evaluation) version of the RANSAC al- gorithm, R-RANSAC, is introduced. Computational savings are achieved by typically evaluating only a fraction of data points for models contaminated with outliers. The idea is implemented in a two-step evaluation procedure. A mathematically tractable class of statistical preverification tests for test sam- ples is introduced. For this class of preverification test we derive an approx- imate relation for the optimal setting of its single parameter. The proposed pre-test is evaluated on both synthetic data and real-world problems and a significant increase in speed is shown. |
Year | Venue | Keywords |
---|---|---|
2002 | BMVC | robust estimator,computer vision,synthetic data |
Field | DocType | Citations |
Computer vision,Pattern recognition,RANSAC,Computer science,Artificial intelligence | Conference | 49 |
PageRank | References | Authors |
4.43 | 10 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jiri Matas | 1 | 335 | 35.85 |
Ondrej Chum | 2 | 5677 | 330.20 |